Therkelsen / MUSE

Muscular Signal Processing for Exoskeleton Control (MUSE)

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Muscular Signal Processing for Exoskeleton Control (MUSE)

Authored by Simon Chris Vinkel and Thomas Therkelsen

Updated December 2023

Introduction

This is the code stack for the MUSE bachelor's project written by Simon Chris Vinkel and Thomas Therkelsen. To read the full thesis, follow this link.

The stack is not quite as polished as we would have liked, that may or may not change in the future, if we continue working on it.

Prerequisites

  1. You'll need a POW-EXO exoskeleton, refer to this paper for more info.
  2. Windows 10/11 PC, as the sensor device used does not run on Ubuntu.
  3. Eliko Quadra device with EIM electrodes, check this link for more info.
  4. Integrated or USB webcam for kinematic angle tracker.

Setup MUSE project

  1. Install Visual Studio

  2. Clone the repo by running the following in your shell: git clone https://github.com/Therkelsen/MUSE.git

  3. Install drivers for Eliko device

    1. From MUSE run Assets/Eliko/Drivers/QuadraUSBDriverInstaller_v1.1.EXE
    2. From MUSE run Assets/Eliko/Drivers/RunTime Installer v20/install.exe
  4. Connect the Eliko device with EIM electrodes

  5. From MUSE open Code/PicometerReader.sln

  6. Make sure you're using the x86 compiler when running or debugging, otherwise it won't build!

  7. When you've run the program, create a throwaway sample (S, K, then Y and Enter) so the startup sequence of the Eliko device doesn't cause a mismatch in time. Now you're ready to record.

Setup Visualizer and Angle Tracker

  1. Install Python 3.11.6

  2. Install requirements

    1. From MUSE run $ pip install -r requirements.txt
  3. Run the visualizer

    1. From MUSE run $ py Code/Visualizer/visualizer.py
    2. It updates the figure once every second, so if you run the MUSE program and gather new data, it will update the figure accordingly.
  4. Run the angle tracker

    1. From MUSE run $ py Code/angle_tracker/angle_calculator.py
    2. Once the program is running, press S to start recording data, K while recording to save, and Q to exit the program.

Setup hotkeys to start and stop recording

  1. Install MacroRecorder

  2. Install PhraseExpress

  3. Open MacroRecorder

    1. Open start_recording.mrf located in Assets/Macros
    2. Click Send to PhraseExpress in the toolbar
    3. Select Hotkey as your trigger in the dropdown, then configure a hotkey combination to start the macro. We used Ctrl + Shift + F8
    4. Repeat for stop_recording.mrf, for that we used Ctrl + Shift + F9
    5. Click File and Save
    6. You may close PhraseExpress, just make sure it's running in the background still so it can register the hotkey combinations.

Record samples

  1. Following Experimental protocol.pdf located in Assets, prepare for sample recording.

  2. Once you have everything prepared, use the start_recording macro hotkey to start recording a sample, and the stop_recording macro hotkey to stop recording.

  3. NOW you're fully ready to record! Great success!

About

Muscular Signal Processing for Exoskeleton Control (MUSE)

License:MIT License


Languages

Language:Jupyter Notebook 65.7%Language:Python 20.7%Language:C++ 11.4%Language:C 1.5%Language:MATLAB 0.6%